This application describes research and development in biomedical computing in the areas of biomedical imaging, neurobiology and cognitive science, as described in NIH solicitation PAR- 118. The objective is to develop novel computational technologies and methods to control cognitive and behavioral training tasks using information from real time functional magnetic resonance imaging (fMRI). This research will test the hypothesis that using this technology, it is possible to train subjects specifically to activate spatially localized brain regions using real time fMRI. Initial pilot experiments suggest that in the near-term this technology will allow conceptually new approaches to understanding brain function, and will improve the speed and quality of fMRI data collection for research purposes. Longer term (beyond this application), this technology may have implications for the understanding or treatment of CNS disease. This development will proceed through the following specific aims: [unreadable] 1) Complete and test a prototype system for real time control of cognitive and behavioral tasks and measurement using real time fMRI. The prototype will be based upon our existing system, and will be cross-platform compatible, and scalable for broad use. [unreadable] 2) Demonstrate proof of concept by using targeted cognitive exercises paired with real time fMRI information to train subjects to increase the level of activity in localized brain regions. [unreadable] 3) Develop novel experimental control paradigm(s) using brain activation-based behavioral control. [unreadable] Gathered data will be used to assess utility and feasibility of development of this technology against specific milestones. [unreadable] [unreadable]